Influenza virus infections are a significant clinical problem in the United States and the possibility that highly virulent influenza virus strains will emerge in the future is a major concern. Since current humoral vaccines are unsatisfactory, there is considerable interest in developing vaccines that emphasize cell-mediated immunity. It has been established that CD8+ T cells play a critical role in the control of influenza virus infections through the lysis of infected lung epithelial cells and the production of antiviral cytokines, However, the mechanisms underlying the induction of these T cell responses are only poorly understood. In the current proposal, this issue will be addressed by determining the factors that regulate the specificity of primary and secondary T cell responses to two influenza virus epitopes (NP[366-374]/Db and PA[224-233]/Db), Previous studies have shown that whereas T cells specific for both of these epitopes are present in equivalent numbers in the lung during the primary response, T cells specific for the NP[366-374]/Db epitope dominate the secondary response. Data are presented showing that these epitopes are differentially expressed in different types of antigen presenting cells. Thus, dendritic cells express both the NP[366-374]/Db and PA[224-233]/Db epitopes, whereas other cell types tested expressed only the NP[366-374]/Db epitope. Based on these observations, it is hypothesized that the changing specificity of T cell response to primary and secondary influenza virus infections in vivo reflects differential antigen presentation. This hypothesis, and the impact of differential antigen presentation on antiviral immunity, will be investigated through the following two specific aims. In Aim 1, the mechanisms underlying changing patterns of immunodominance during influenza virus infection will be investigated. Studies will identify the subsets of antigen presenting cells involved in the response, their capacity to traffic antigen to the lymph nodes, and their capacity to stimulate naive versus memory T cells. In Aim 2, the impact of NP[366-374]/Db and PA[224-233]/Db immunodominance patterns on antiviral immunity will be determined. Studies will focus on the capacity of PA[224-233]/Db -specific T cells to recognize virally infected lung epithelial cells and the impact of PA vaccination on antiviral immunity. Together, these studies will provide new insights into the mechanisms underlying T cell responses in vivo and will have significant implications for the development of vaccines that promote cellular immunity to pulmonary infections.